MEASUREMENT ERROR

Testing kids for “grit” is a big mistake, says the world’s foremost authority on it

This autumn, a handful of California school districts will start testing kids on “non-cognitive” skills like grit, mindset, and self-control, the New York Times reported today (Mar. 1).

Changes to a federal education law require states to assess schools on at least one non-academic measure. Research has shown that social-emotional traits like self-control and perseverance can be better predictors of success than standardized test scores, or other purely academic measures.

But Angela Duckworth, perhaps the world’s foremost authority on grit and the author of much of the research on the topic, says this is a big mistake.

“I do not think we should be doing this; it is a bad idea,” Duckworth told the Times.

Duckworth, a professor at the University of Pennsylvania, has shown that grit, defined by her as the “tendency to sustain interest in and effort toward very long-term goals,” is a good predictor of success in everything from surviving the grueling trials of military academies to reaching the final rounds of spelling bees and simply graduating from public high school.

She has also shown than self-control can predict report-card grades, and changes in these grades over time, better than measured intelligence.

Naturally, understanding how to improve kids’ grit is of interest to educators and parents. Duckworth thinks that it can can be taught, and that it can grow over time. This is similar to the work of Stanford researcher Carol Dweck, who discovered that kids learn more when they believe intelligence is malleable, and that a growth mindset can be developed over time.

Duckworth has previously outlined her misgivings about the measurement of non-cognitive abilities she thinks are critically important. Testing these traits will be difficult, in part because there are so many of them (empathy, self-control, mindset, zest) and there are a range of definitions for them. Another major obstacle is the tests are faulty: self-reported surveys and teacher questionnaires present a host of problems, including students’ and teachers’ biases. As she wrote, along with David Scott Yeager from the University of Texas at Austin:

Our claim is not that everything that counts can be counted or that everything that can be counted counts. Rather, we argue that the field urgently requires much greater clarity about how well, at present, it is able to count some of the things that count.

Here is Duckworth’s attempt at a grit test, which can produce some pretty weird results.

Some New York Times readers comments were appropriately skeptical of California’s grit-testing plans. “So when do parents get to sign their kids up for ‘grit building’ classes?,” asked one, hardly a stretch when parents now hire tutors for nursery school interviews. Said another, “Tests for joy? The very definition of an oxymoron.” More than a few noted also noted that schools seem to be giving rewards to kids for behavior once considered standard (conscientiousness, self-control).

Testing has been on the rise in the United States for some time, much to the chagrin of many parents who say it is stressing their kids out and does nothing for their academic achievement. More testing for more ambiguous measures hardly seems like the solution.

home our picks popular latest obsessions search